AI tool helps dermatologists decide if a skin lesion could be melanoma

Is this trial a fit for you?

A plain-English reading of the study's rules. Only the study team can confirm whether you qualify.

What this trial studies

This Phase 3 trial evaluates whether the DEXI artificial intelligence decision-support system improves melanoma screening in dermatology clinics. DEXI analyzes images and information to flag lesions that may be melanoma when used with the DermaGraphix Visualizer. Participants are adults undergoing in-person visits at participating sites, and clinicians use DEXI as part of real-time decision-making. The study looks at changes in biopsy decisions and melanoma detection rates compared with standard care. Lead sponsor is Memorial Sloan Kettering Cancer Center, with NIH and Canfield Scientific as collaborators, and data collection occurring at Stanford for the imaging component.

Why it matters

Who can join: the full rules

Inclusion Criteria:

* Age 18 years or older at the time of scheduling the office visit.
* Presenting for an in-person office visit with an investigative provider at a participating clinical site during the study period. Both referral visits and recurring skin checks are eligible.
* Visit must have a mapped ICD-10 code with at least one of the following:

ii. iii. C43 - Malignant melanoma of skin C44 - Other and unspecified malignant neoplasm of skin D03 - Melanoma in situ iv. D04 - Carcinoma in situ of skin v. D18.0 - Hemangioma vi. D22 - Melanocytic nevi vii. D23 - Other benign neoplasms of skin viii. D48.5 - Neoplasm of uncertain behavior of skin ix. D48.9 - - Neoplasm of uncertain behavior, unspecified x. D49.2 - Neoplasm of unspecified behavior of bone, soft tissue, and skin xi. I78.1 - Nevus, non-neoplastic xii. L57 - Skin changes due to chronic exposure to nonionizing radiation xiii. L73.8 - Other specified follicular disorders xiv. L81.4 - Other melanin hyperpigmentation xv. L82 - Seborrheic keratosis xvi. Q82.5 - Congenital non-neoplastic nevus xvii. Z12.83 - Encounter for screening for malignant neoplasm of skin xviii. Z80.8 - Family history of malignant neoplasm of other organs or systems xix. Z85.82 - Personal history of malignant neoplasm of skin xx. Z86.006 - Personal history of melanoma in-situ xxi. Z86.007 - Personal history of in-situ neoplasm of skin

* Dermatologist conducting the visit must have undergone training for the how to use and interpret the CDS.

Exclusion Criteria:

* Telehealth-only, home, inpatient, emergency department, or procedure-only encounters without an in-office visit.
* If any image data captured on the patient were known to have been used for training or validating DEXI.

Where it's running

• Memorial Sloan Kettering at Basking Ridge (All protocol activities)
  Basking Ridge, New Jersey, United States Recruiting
• Memorial Sloan Kettering Monmouth (Consent only)
  Middletown, New Jersey, United States Recruiting
• Memorial Sloan Kettering Bergen (Consent Only)
  Montvale, New Jersey, United States Recruiting
• Memorial Sloan Kettering Cancer Center @ Suffolk-Commack (Consent only)
  Commack, New York, United States Recruiting
• Memorial Sloan Kettering Westchester (All Protocol Activities)
  Harrison, New York, United States Recruiting
• Memorial Sloan Kettering Cancer Center (All Protocol Activites)
  New York, New York, United States Recruiting

Study contacts

• Principal investigator: Veronica Rotemberg, MD — Memorial Sloan Kettering Cancer Center
• Study coordinator: Veronica Rotemberg, MD
• Email: rotembev@mskcc.org
• Phone: 646-608-2367

Similar studies

Other studies for the same condition or treatment. Recruiting studies are listed first.

• Fluctuational ultrasound imaging to help diagnose liver hemangioma

  This project tests whether a special ultrasound method called fluctuational imaging can help doctors identify liver hemangiomas in adults who have focal liver lesions of 1 cm or larger.

  Recruiting Yonsei University · Seoul, Seodaemun-gu · NCT07418294

• Early skin tumor detection with telemedicine in primary care

  This project will test whether a telemedicine tool used by general practitioners can help detect suspicious skin lesions earlier in people with fair skin in the Czech Republic.

  Recruiting at 2 sites Institute of Health Information and Statistics of the Czech Republic · Brno and 4 other locations · NCT07349940

• Confocal imaging and genetic testing of atypical moles and early melanoma

  This project will try to see if noninvasive confocal images of suspicious moles match genetic changes from sequencing in adults who are having the lesions removed.

  Recruiting University of Modena and Reggio Emilia · Modena, Mo · NCT07277920

• Understanding diagnostic and prognostic factors in melanoma

  This study is trying to find out how different types of skin growths, like melanoma and benign moles, affect patients by looking at their medical history and collecting blood samples.

  Recruiting at 4 sites Region Skane · Helsingborg, Skåne County and 3 other locations · NCT05446155

• Understanding the genetic causes of moles

  This study is trying to find out the genetic changes that cause moles by looking at DNA from skin samples and blood to better understand these common skin growths.

  Recruiting University of California, Davis · Sacramento, California · NCT03054584

• Comparing pulsed dye laser and timolol for treating hemangiomas in newborns

  This study is testing whether a special laser or a cream can help treat hemangiomas in newborns better than just watching them grow.

  Recruiting Massachusetts General Hospital · Boston, Massachusetts · NCT01873131

How to take part

1. Review the eligibility criteria above with your treating physician.
2. Check the official listing on ClinicalTrials.gov for the latest contact details and recruitment status.
3. Contact the study coordinator or principal investigator to ask for pre-screening. Pre-screening is free and never obligates you to enroll.